II. EXPERIMENTAL

Fingolimod hydrochloride was commercial reagent, purchased from AbMole BioScience, and was used as-received. The white powder was packed into a 1.5 mm diameter Kapton capillary, and rotated during the measurement at ~50 cycles s⁻¹. The powder pattern was measured at 295 K at beam line 11-BM (Lee et al., Reference Lee, Shu, Ramanathan, Preissner, Wang, Beno, Von Dreele, Ribaud, Kurtz, Antao, Jiao and Toby2008; Wang et al., Reference Wang, Toby, Lee, Ribaud, Antao, Kurtz, Ramanathan, Von Dreele and Beno2008) of the Advanced Photon Source at Argonne National Laboratory using a wavelength of 0.413 906 Å from 0.5° tο 50°2θ with a step size of 0.001° and a counting time of 0.1 s step⁻¹. The pattern was indexed on a primitive monoclinic unit cell having a = 7.1414, b = 5.9600, c = 49.5275 Å, β = 91.046̊, V = 2106.69 Å³, and Z = 4 using DICVOL06 (Louër and Boultif, Reference Louër and Boultif2007). The space group was indicated to be P2₁/n by EXPO2013 (Altomare et al., Reference Altomare, Cuocci, Giacovazzo, Moliterni, Rizzi, Corriero and Falcicchio2013). A reduced cell search in the Cambridge Structural Database (Allen, Reference Allen2002) yielded 11 hits, but no crystal structure for fingolimod hydrochloride.

A fingolimod molecule was built and its conformation optimized using Spartan ’14 (Wavefunction, 2013), and saved as a mol2 file. This file was converted into a Fenske–Hall Z-matrix file using OpenBabel (O'Boyle et al., Reference O'Boyle, Banck, James, Morley, Vandermeersch and Hutchison2011). This molecule and a chlorine atom were used as fragments to solve the structure using FOX (Favre-Nicolin and Černý, Reference Favre-Nicolin and Černý2002). Positions of the active hydrogens were deduced by an analysis of potential hydrogen bonding.

Rietveld refinement was carried out using GSAS (Larson and Von Dreele, Reference Larson and Von Dreele2004). Only the 0.8°–20° portion of the pattern was included in the refinement, and the region 1.3°–1.8° (which contained an unindexed impurity peak of ~2% relative intensity) was excluded. The C22-H31 C₆H₄ phenyl ring was refined as a rigid body. All other non-hydrogen bond distances and angles were subjected to restraints, based on a Mercury/Mogul Geometry Check (Bruno et al., Reference Bruno, Cole, Kessler, Luo, Motherwell, Purkis, Smith, Taylor, Cooper, Harris and Orpen2004; Sykes et al., Reference Sykes, McCabe, Allen, Battle, Bruno and Wood2011) of the molecule. The Mogul average and standard deviation for each quantity were used as the restraint parameters. The restraints contributed 0.92% to the final χ ². Isotropic displacement coefficients were refined, grouped by chemical similarity. The hydrogen atoms were included in calculated positions, which were recalculated during the refinement. The U _iso of each hydrogen atom was constrained to be 1.3× that of the heavy atom to which it is attached. The peak profiles were described using profile function #4 (Finger et al. Reference Finger, Cox and Jephcoat1994; Thompson et al. Reference Thompson, Cox and Hastings1987), which includes the Stephens (Reference Stephens1999) anisotropic strain broadening model. Unusually, the S/L profile term was refined to obtain the best fit; it refined to a value of 0.002 249(7), compared with the instrument value of 0.0011. The background was modeled using a three-term shifted Chebyshev polynomial and a six-term diffuse scattering (Debye) function, to model the scattering from the Kapton capillary and any amorphous component of the sample. Given the expected platy morphology of the crystals, a 4th-order spherical harmonic preferred orientation correction was included in the model. The final refinement of 96 variables using 18 736 observations (18 699 data points and 37 restraints) yielded the residuals R _wp = 0.0697, R _p = 0.0569, and χ²  = 1.166. The largest peak (1.30 Å from N12) and hole (1.55 Å from C1) in the difference Fourier map were 0.24 and −0.19 e (Å⁻³), respectively. The Rietveld plot is included as Figure 2. The largest errors are in the shapes of some of the low-angle peaks, which may indicate subtle changes in the specimen during the measurement.

Figure 2.

(Color online) The Rietveld plot for the refinement of fingolimod hydrochloride. The red crosses represent the observed data points, and the green line is the calculated pattern. The magenta curve is the difference pattern, plotted at the same vertical scales as the other patterns. The vertical scale has been multiplied by a factor of 5 for 2θ > 2.0°, and by a factor of 20 for 2θ > 10.0°. The 1.3°–1.8° excluded region (which contains the impurity peak) is not included.

A density functional geometry optimization (fixed experimental unit cell) was carried out using CRYSTAL09 (Dovesi et al., Reference Dovesi, Orlando, Civalleri, Roetti, Saunders and Zicovich-Wilson2005). The basis sets for the H, C, N, and O atoms were those of Gatti et al. (Reference Gatti, Saunders and Roetti1994), and the basis set for Cl was that of Apra et al. (Reference Apra, Causa, Prencipe, Dovesi and Saunders1993). The calculation used 8 k-points and the B3LYP functional.

III. RESULTS AND DISCUSSION

The refined atom coordinates of fingolimod hydrochloride are reported in Table I, and the coordinates from the density functional theory (DFT) optimization in Table II. The U _iso of the atoms in the octyl side chain are not especially large, providing no indication of disorder of the side chain. The root-mean-square deviation of the non-hydrogen atoms is 0.60 Å, and the maximum deviation is 1.97 Å, at O8 (Figure 3). The differences indicate difficulty in refining the conformation of the octyl side chain. Most notable is the difference between the orientation of the hydroxyl group O8–H9 in the refined and optimized structures. Re-starting the refinement from the DFT-optimized structure leads to the same refined model reported here. We believe that the DFT structure is correct, as it results in a more chemically reasonable hydrogen-bonding pattern, but report both structures here so that the reader can choose his/her preferred one. The discussion of the geometry uses the DFT-optimized structure. The asymmetric unit (with atom numbering) is illustrated in Figure 4, and the crystal structure is presented in Figure 5.

Figure 3.

(Color online) Comparison of the GSAS-refined and DFT-optimized structures of fingolimod hydrochloride. The Rietveld-refined structure is colored red, and the DFT-optimized structure is in blue.

Figure 4.

(Color online) The molecular structure of fingolimod hydrochloride, with the atom numbering. The atoms are represented by 50% probability spheroids.

Figure 5.

(Color online) The crystal structure of fingolimod hydrochloride, viewed down the a-axis.

Table I.

Rietveld refined crystal structure of fingolimod hydrochloride.

TABLE II.

Density functional theory optimized (CRYSTAL09) crystal structure of fingolimod hydrocloride.

The powder pattern of the form characterized here appears to correspond to that of form α (the room temperature polymorph) claimed in the US Patent 2013/0281739 A1 (Shrawat et al., Reference Shrawat, Veereshappa, Singh and Purohit2013). The patterns of forms β and μ differ significantly.

All bond distances and angles, and most torsion angles fall within the normal ranges indicated by a Mercury Mogul Geometry Check. The torsion angles O10–C3–C2–C16, C24–C25–C32–C33, C26–C25–C32–C33, and C36–C35–C34–C33 are flagged as unusual, but they lie on the tails of the normal distributions of these torsion angles. The hydrogen bonds certainly contribute to the orientation of the hydroxyl groups.

Conformational analysis of the fingolimod cation (Hartree–Fock/3-21G/water) using Spartan ’14 suggests that the solid state conformation is at least 24 kcal mol⁻¹ higher in energy than the minimum-energy conformation. The differences lie mainly in the orientations of the phenyl ring and the polar end of the molecule. Apparently hydrogen bonding is important in determining the solid state conformation. An analysis of the contributions to the total crystal energy using the Forcite module of Materials Studio (Accelrys, 2013) suggests that bond, angle, and torsion distortion terms contribute little to the crystal energy. The energy appears to be dominated by van der Waals and electrostatic contributions, which in this force-field-based analysis include hydrogen bonds. The hydrogen bonds are better analyzed using the results of the DFT calculation.

The structure consists of a “lipid bilayer” packing. The polar ends of the molecules make O–H···Cl and N–H···Cl hydrogen bonds (Table III) to the chloride anion, and the octyl side chains pack adjacent to each other. The hydrogen bonds form three types of chains with graph sets (Etter, Reference Etter1990; Bernstein et al., Reference Bernstein, Davis, Shimoni and Chang1995; Shields et al., Reference Shields, Raithby, Allen and Motherwell2000) C1,2(7), C1,2(7), and C1,2(8). The result is a complex chain of hydrogen bonds parallel to the b-axis. The ammonium hydrogen H15 does not participate in hydrogen bonds.

TABLE III.

Hydrogen bonds in fingolimod hydrochloride.

The Bravais–Friedel–Donnay–Harker (Bravais, Reference Bravais1866; Friedel, Reference Friedel1907; Donnay and Harker, Reference Donnay and Harker1937) morphology suggests that we might expect platy morphology for fingolimod hydrochloride, with {001} as the principal faces. A fourth-order spherical harmonic preferred orientation model was included in the refinement; the texture index was 1.083, indicating that preferred orientation was significant even for this rotated capillary specimen. The powder pattern of fingolimod hydrochloride has been submitted to ICDD for inclusion in future releases of the PDF.